The purpose of this proposal is to conduct a full day national conference to improve the design, statistical analysis and interpretation of long term trials in bipolar disorders. The program will be promoted to investigators across the United States in bipolar disorders. The conference will provide a venue to bring together leaders in bipolar methodologies in diagnosis, study design, symptomatic characterization and strategies of effective long term care. Bipolar disorder has come to be understood as symptomatically and functionally complex, with most patients symptomatic approximately half of the time over long periods even in the context of optimally applied professional interventions. Biostatisticians have developed more rigorous and complex statistical techniques to deal with these problematic realities. These techniques deal with incomplete data, provide stronger power to detect significant effects, and yield information on real world outcomes, not just outcomes in uncommonly straightforward patients with bipolar disorders. The procedures can also improve reliability of measurements and increase study power. The recently initiated NIMH Bipolar Trials Network investigators will be able to apply substantial components of the recommendations in interventions studies of bipolar disorders. Similarly with a substantial proportion of bipolar investigators in the U.S. attending the conference; these benefits will have broad impact on the design and generalizability of long term studies on improving treatments for bipolar disorder. The design recommendations from the conference can also strengthen approaches to the study of new treatments for bipolar disorders. A major approach to improving outcomes in bipolar disorder entails more efficient application of long term interventions in regular clinical practice settings. This conference of leading biostatisticians and clinical investigators on bipolar disorder is structured to provide recommendations and design approaches which can advance design and statistical methodologies for bipolar disorder for purposes of reducing illness burden, improving sustained functional outcomes and contributing to development of novel treatments. [unreadable] [unreadable] [unreadable]